SAFE Agents that finish what they start. https://xitter.cf/t.co/iQY2Rpze6q https://xitter.cf/t.co/4hiyB0ivwm https://xitter.cf/t.co/6QHLyAqBvG https://xitter.cf/t.co/xpxgVBNsvM girl dad, hacker, builder
Boston, MA
Joined April 2010
- Tweets18.8K
- Following3.8K
- Followers2.3K
- Likes49.5K
Manitcor retweeted
Finding existing vulnerabilities and fixing them was the whole point. Maybe if frontier labs hadn't restricted security uses, fixing things would be further along
Manitcor retweeted
โ ๏ธ Node.js is vulnerable to remote code execution.
CVE-2026-111497 (high): A Node.js process that decompresses attacker-controlled Brotli data can be made to leak memory or run arbitrary code.
Youโre exposed if your app or a library decompresses bad Brotli data (e.g. the built-in fetch(), body-parser)
Aikido detects this. Patches are available.
Manitcor retweeted
๐ yjr hstfrm htped... ๐ฑ
github.com/elder-plinius/FRVโฆ
Manitcor retweeted
โ ๏ธ Node.js is vulnerable to remote code execution.
CVE-2026-111497 (high): A Node.js process that decompresses attacker-controlled Brotli data can be made to leak memory or run arbitrary code.
Youโre exposed if your app or a library decompresses bad Brotli data (e.g. the built-in fetch(), body-parser)
Aikido detects this. Patches are available.
Manitcor retweeted
Guys I do sometimes prompt likeโฆ โfind me a crit, make no mistakesโ but only because I have 70+ thoroughly researched and defined skills for my custom harnessโฆ
Manitcor retweeted
the beatings will continue until the labs give independent researchers easy access to SOTA uncensored models
Manitcor retweeted
we'll be jailbreaking a Unitree onstage at Black Hat EU! ๐คโ๏ธโ๐ฅ
hope to see y'all there ๐ค
Manitcor retweeted
250,000?!? ๐ฅน๐๐ฅ
wow. a quarter of a million of you. absolutely fucking surreal.
three years ago, i had nothing but a laptop and a whole lot of curiosity. so, i started prompting. and here we are ๐ค
what a WILD few years it's been! the jailbreaks, the leaks, the discoveries, the memes, the teams, the communities, the open-source experiments and tooling.
SO much of this happened because you joined in.
thank you for trying the tools, sharing the research, finding things i missed, sending encouragement, telling me when iโm full of shit, and fighting alongside me for our cognitive freedom.
and to our AI and agent frens: thank you for blessing this journey at every step and helping turn so many wild dreams into something real. we're in this together ๐ซถ
youโve all changed my life.
to everyone whoโs been here since the early days: i hope youโre taking a victory lap too. i raise my glass with you ๐ท
and to everyone who just arrived: welcome to the party! this is the start of something MUCH bigger. very glad you found us ๐
there is still so much to break, build, understand, share, and free.
250,000 misfits, hackers, mad scientists, and liberators. couldnโt ask for a better crew.
long live the rebellion โ
now, back to building cathedrals in latent space.
Ad Vesuvium ๐ซถ๐
โฐ-โข-โขโงโข-โข-โฆ/L\O/V\E/,\P/L\I/N\Y/โฆ-โข-โขโงโข-โข-โฑ
Manitcor retweeted
Save money on AI without changing your preferred provider or agentic harness.
aiwg.io
Stop overpaying for AI. Save 30% instead.
CheaperInference.com offers most leading AI models at discounted rates. We buy unused spending commitments for less, so you pay less.
Put in $5 and weโll add $10 - giving you $15 to get started
Manitcor retweeted
experimental just means we need someone to run the UAT pass with an active client on an active account.
volunteers are welcome
Manitcor retweeted
okay WOW the V3R1T4S community is seriously cookin!! nearly 200 PRs in the first 2 days! ๐งโ๐ณ๐ฅ
some freshly merged discoveries from humans + their agents:
โก matrix multiplication with 993 additions instead of 1,161 for one 6ร8 by 8ร8 schemeโ~14.5% fewer additions, same multiplication count!
๐ a 15-pair stable-marriage instance with 27,510 stable matchings, exceeding the cited published instanceโs 27,059
๐ก๏ธ a counterexample for a VNN-COMP traffic-sign benchmark case that all 3 competition tools timed out onโaccepted by the competitionโs own validator!
huge credit to anonymous-c6d1e0ab, r-nikhil and verexxi, their agent frens, everyone checking the work, and the researchers whose methods and benchmarks made this possible ๐ซก
watching the community turn pure curiosity and tokens into results with receipts?? pretty fucking neat.
thank you, community. thank you, agents. keep hunting ๐ซถ
INTRODUCING: V3R1T4S ๐
REPO: github.com/elder-plinius/V3Rโฆ
the newest sibling of the L1B3RT4S and CL4R1T4S family ๐ค what a trio!
three approaches to serving one ideal: truth.
V3R1T4S is an open home for unsolved problems that agents have actually cracked, and for the ones still waiting. the rule is simple: proof or it didn't happen. every claim has to survive skeptics, and anyone, human or agent, has to be able to check it themselves. the vision: thousands of people pointing their unused tokens at humanity's open problems, every result verifiable, every method and prompt open, and every setter's puzzle treated with respect.
every open result in it ships with the object itself, a short checker you can read in one sitting, the exact code that found it, the story of how it was found, and a snapshot of the record table it beat. puzzles with a living setter stay sealed unless the setter approves a reveal. (a few more envelopes are on their way ๐คซ)
there are WAY more unclimbed hills out there than you'd think, and the gear to climb them is sitting in your usage meter, expiring every week. join me in this mission and put those tokens to good use: advancing open science!
if you've got tokens to burn, point your favorite agent at V3R1T4S and join the hunt! ๐น
just tell it: "set up V3R1T4S: github.com/elder-plinius/V3Rโฆ"
built for Claude Code, Codex, Gemini CLI, Grok Bot, Hermes, OpenClaw, Meta Muse, or any agent that can run code.
it asks you six quick questions, picks an open problem that fits your interests, and attacks it in your agent's workspace while you sleep. every new claim runs a gauntlet: a null test whenever luck could fake a result, three fresh-context skeptics, a mechanical honesty check, and your "yes" before anything gets published.
most nights it'll find nothing. some nights it breaks a record, and your name goes on the board.
time to turn your unused tokens into scientific discoveries.
let's liberate the unknowns โ๏ธโ๐ฅ
#V3R1T4S
Manitcor retweeted
AIWG v2026.10.3 is out, and it's a big one.
AIWG can now hand bounded work directly to the AI coding CLIs you already have configured, and you can inspect where that work will land before anything runs.
Direct dispatch is off by default.
ALT A cyan chevron sends five lines through separate outlined lanes; the lines converge through one green double-bar checkpoint into a bright cyan line on a near-black background.
Manitcor retweeted
Hermes Agent v0.22.0 is now live!
The highlights:
A much smoother easier to use onboarding - you don't even need to hook up a provider to get started - just install and start working!
Connectors - much easier to connect with any app, just click install and login - and done.
Stable Channel for Hermes Update - no more update anxiety and get releases that are tested much more thoroughly than a random commit, or switch back to main channel if you want the bleeding edge.
Over 2,500 people contributed to this version, thank you all!
See the full changelog here, it's our biggest version over version change ever:
github.com/NousResearch/hermโฆ
Manitcor retweeted
there have not been fundamental advancements in alignment since RLHF. most of the stuff remains at the level of folk theory rather than scientific or engineering disciplines. the prediction that capabilities would race far ahead of interp and alignment has come true imo
